* feat: split ai agent memory into agent policy, run memory id and step history
* fix: scope string memory ids to workspace and flow, keep nested tool history inputs
* chore: update sqlx cache for the flow context query
* docs: describe memory id scoping as collision-free rather than isolated
* chore: regenerate openflow json after merging main
* fix: offer no memory id for legacy manual memory, document linked history inputs
* fix: seed provided messages from legacy manual memory and hide its note once set
* fix: bypass memory when a provided messages expression evaluates to null
* fix: require a user message when provided messages are empty
* chore: keep the empty messages comment within the line width
* docs: name the history inputs wherever linked steps list their flow-local inputs
* docs: keep the memory storage path on one line
* feat: managed memory with an inherited or custom memory id per step
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fix: list a custom memory id in the test run form and name where an inherited one comes from
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fix: keep memory id out of the add-field menu and drop the memory id telemetry
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fix: keep legacy auto memory without an id working after an untouched redeploy
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fix: rename step messages to previous_messages and address review
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* style: rewrap comments and docs lines lengthened by the previous_messages rename
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* refactor: read agent memory as either a legacy shape or the current one
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fix: name the memory setting in ignored-input notes and keep conversions honest
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fix: keep a legacy memory count unset on open and read a cleared count as off
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: address review on cleared test history and zero-count memory
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: drop flow-local keys from a linked agent resource before interpolating it
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: keep a linked resource's own inputs as fallbacks and note ignored history on image runs
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: restore the linked agent draft tests and log ignored history on every image run
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: resolve the one-of variant from the value when the selected one leaves the list
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: treat zero-count managed memory as off when enabling chat mode and shorten comments
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: stop requiring user_message in the openflow agent contract when previous messages are the prompt
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
* feat: mark test flow conversations apart from deployed ones and allow renaming a chat
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: keep the conversation kind across refreshes and reject NUL titles
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: ignore conversation lists for a kind no longer selected
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: start a fresh conversation listing when the kind changes on a later page
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: refuse sending into a conversation of the other kind
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: refuse cross-kind conversation continuations on the server
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: keep the conversation filter unavailable while an answer runs
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* refactor: render the session chat model menu from a shared ChatModelSettings config
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* feat: pick the flow chat's model and thinking from the provider fields the flow exposes
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: name only the thinking level the flow run will send on the model button
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: let the flow chat take a typed model id and keep a shared thinking input editable
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: promote a flow input to the model button only where its control can edit it
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: drop any reasoning token the chosen model rejects before a flow chat run
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* feat: keep flow inputs and seed the agent when chat mode is enabled
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: leave a linked agent's memory and streaming to the agent when enabling chat mode
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* feat: collect flow conversations and agent memory when their last message goes
* fix: lock the conversation lookup so a new turn orders against its cleanup
* fix: let concurrent turns recreate a collected conversation without conflicting
* feat: render the flow chat through the shared session chat components
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: settle the flow chat queue on the turn's outcome and guard IME enter
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: keep copilot free-tier banners and deploy-locked sends out of the flow chat
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* fix: skip expiry notifications for app embed and SDK tokens
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* refactor: share app token label prefixes between mint sites and the check
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fix: skip expiry alerts for impersonation and test-connection tokens
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
* feat: show flow step detail as a page inside the graph tab on narrow detail layouts
* fix: pad the detail step tab, wrapped header row and raise the tabbed layout breakpoint
* fix: keep the step header pinned and switch to triggers on every trigger node tap
* feat: badge chat-input flows on the home list
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: keep a malformed draft value from aborting the runnables list
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: only a JSON boolean marks a draft flow as chat-enabled
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* fix(apps): re-check access in place after a password sign-in
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* style: drop redundant comment in the password sign-in hand-back
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
* feat: stream reasoning summaries in AI agent Responses API steps
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fix: retry without a reasoning summary a strict gateway rejects
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fix: request the reasoning summary whether or not the step streams
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* feat: return the OpenAI reasoning summary in the agent step's reasoning
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* fix: retry without a reasoning summary a gateway rejects with 422
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
* fix: walk the whole fork ancestry for app installations and fork conflicts
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs: describe the fork-conflict gate as ancestor-wide
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* chore: update ee-repo-ref to d252afcc80e77fcc4f9a2a346b80908c8605a6c0
This commit updates the EE repository reference after PR #803 was merged in windmill-ee-private.
Previous ee-repo-ref: 5f68c8c351ffc92feccffe69a857b60be376464e
New ee-repo-ref: d252afcc80e77fcc4f9a2a346b80908c8605a6c0
Automated by sync-ee-ref workflow.
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
* feat: list flows that link a saved agent and flag broken agent links
* feat: rename saved agents from the agent editor and repoint the flow
* fix: show an unreadable linked agent as not accessible, not missing
* fix: address review nits on agent rename and missing-agent state
* fix: open content search above modals and keep Escape for it
* fix: register content search on the opener's overlay stack
* docs: scope the global search z-index comment to the bases it clears
* refactor: show linked agents' rename warning as for scripts and flows
* fix: keep the failed-lookup rename warning to resources
* fix: dispatch workflow-as-code tasks from a deployed flow's inline step
* fix: give a workflow-as-code task its own result-cache key
* fix: key a cached workflow-as-code task on its name and arguments
* fix: hash a cached workflow-as-code task's arguments like any job's
* chore: regenerate system prompts for the task cache_ttl docs
* fix: key a cached workflow-as-code task on its step key, not its name
* fix: key a cached workflow-as-code task on a fingerprint of its code
* fix: keep the task() doc attached to task()
* fix: key a cached inline task on its step key and the workflow input
* docs: cache_ttl has no effect on a taskFlow target
* feat: delete a browser's copy of an AI session past its workspace retention
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: tell the AI session retention only to a member who can reach the workspace
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* docs: keep the retention sweep's design narrative in the docs, not the code
* fix: give the session retention its own route, leaving the status contract alone
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* docs: shorten the retention route comment to its constraints
* docs: name the two clocks in the retention setting, and the deploy window
---------
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
* feat: retention for AI sessions, swept on the object store and in the browser
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: make the retention sweeps retryable and safe against pushes
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: spare other tabs' sessions, reclaim abandoned split pushes
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: sweep under an exclusive session lock, keep the captured user
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* docs: say a tab selecting a session mid-sweep is not held back
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: sweep local sessions only while no other tab has them loaded
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: one tab sweeps at a time, and keeps the switched user's hold
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* test: push the fallback session again before the rotation assertions
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* refactor: keep retention server-side here, move the browser sweep out
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* docs: the retention setting no longer touches browser-local sessions
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* feat: instance object store as fallback for AI session backups
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: fence the instance store sweep by generation, name it by location
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* test: pin that an instance store location tells endpoints apart
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: show the instance storage fallback setting on while it is unset
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: check the generation fence queries at compile time
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
* fix: stop the instance storage fallback once the plan is Pro
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* test: pin that a failed multi-object incremental push leaves the session unlisted
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* test: pick the newest backup generation in the ai sessions test helper
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* feat: back AI sessions up to the workspace object storage
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SG5qEPM6Fmf7VerXS5nnWp
* fix: bind the backup key to the user and pack pushes within the server caps
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SG5qEPM6Fmf7VerXS5nnWp
* fix: keep refused and unavailable marks, one mark per key, stream the flush
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SG5qEPM6Fmf7VerXS5nnWp
* fix: settle only fully sent sessions, keep removals while backups are off, cap pull bodies
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01SG5qEPM6Fmf7VerXS5nnWp
* fix: bound removal marks while backups are off and stale the sync rows instead of dropping them
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: retry a lost lock, cap nested push lists and oversized pieces, drop a stale copy of a chat that outgrew the backup
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: cap pieces per push, size requests in UTF-8, keep a move's removal for an off workspace, disclose the restore counter
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: file a move's removal only once the new copy landed, retire marks through the sync row
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: keep a session marked while deletes are carried over, drop only gone sessions' marks
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: record what a refused flush already stored, stop early when every mark is retired
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: restore past another workspace's removal mark, file a move's removal before its row, bound the first pulled session
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: re-key the backups on workspace key rotation, accept only base64 images, carry a delete on the sync row when its mark cannot be written
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: durable conditional re-key of session backups, re-push on a storage switch
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: fail a push the key rotated under, settle no session split across storages, narrow the re-key module
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: keep a delete filed during a push, bound the pull and re-key listings
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: bound the session listing, mark the store's own user on a write that lands after a user switch
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: list sessions through per-session index markers, hold a session's parts back after a failed one
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: record a rotation on every build, list a session only on the part that completes its push
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: leave an object larger than any push writes unread
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: read each object against its listed size, carry a dirty mark that cannot be written on the sync row
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: note the storages the re-key walk completed on, reach another user's rows on a failed mark, read a head at its cap
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: read the replaced key under its row lock, carry a refused dirty bump on the sync row
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: pull a session that outgrew one answer in pages, imported only whole
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: build a pull page from the smallest keys of the whole listing, stage each page as it lands
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: re-record a key rotated back to, admit earlier-page images, restage over a cut-short restore
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* refactor: delete the backups on key rotation instead of re-keying them
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: end a pull page before an object that grew since the listing, prune what a cut-short restore staged
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: delete the backups before the key commits, skip a planted object whatever its listing says, lock a restore across tabs, prune stale artifact versions
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: keep the backups under a prefix named by the key, delete the previous key's prefix after the commit
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: name the backup prefix by a generation the rotation bumps, never write an older record over a newer one on restore
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: retire a removal only against the storage holding the backup, restart a paged pull whose listing moved
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: fingerprint a pull page before reading it, answer the backup generation apart from the storage identity
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: answer needs_head for a headless session push, prune restaged pieces by id
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: serialize a session's push and removal, open whole pushes with the head, prune only own restores
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: require a head on a whole push, prune before the record lands, restore only under Web Locks
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: incremental pushes ride on a listed session, removals wait for every storage holding a copy
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: a whole push replaces the backup under a per-push token, a pull page is checked after its reads
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: fingerprint a pull page by entity tag and version too
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: a moved session's removal mark names the storages holding the old copy
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: restore a workspace family together, the newest copy of a moved session winning
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: name the marker by the session's move count, abort a family restore a listing failed in
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: list the family again before a restored record lands, require the pull fingerprint
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: list the whole family once per restored workspace, off members included
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: a push split over parts, incremental too, unlists the session until its last part
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: refuse a partial push part that names no push
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: keep a refused bump for a session with no row yet, probe an off workspace again
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: backfill row-carried bumps after a reload, ask an off workspace again on a timer
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: backfill a row for its bumps only when it carries some
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: a backfilled mark that cannot be written counts from the page's counter
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs: say an off workspace is asked again, in the mirror's comments
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* chore: update ee-repo-ref to 1c1dab33563c4907aff8b0da825fb66db60af82a
This commit updates the EE repository reference after PR #796 was merged in windmill-ee-private.
Previous ee-repo-ref: 289b477ca3fc993da06ec09b11c8f55d5e4e39c1
New ee-repo-ref: 1c1dab33563c4907aff8b0da825fb66db60af82a
Automated by sync-ee-ref workflow.
* fix: unlist a session while an incremental push changes more than one object
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Co-authored-by: windmill-internal-app[bot] <windmill-internal-app[bot]@users.noreply.github.com>
* fix(cli): keep the workspace color when settings are synced from git
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs(cli): name the sync direction consistently in the identity-field comments
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(cli): apply the workspace color from settings.yaml only when the file sets one
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* refactor: run the flow chat UI on the windmill-chat sdk
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: structural answer check, poll option and latest run in the chat sdk
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix: count jobless tool rows and the loaded license in the flow chat
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
